MAY 15, 1935
ANALYTICAL EDITION
Conclusions As a reproducible test of color stability of gasoline, exposure to a carbon arc burning "sunshine" carbons using 60 amperes a t 50 volts alternating current is recommended. In the latitude of Chicago, 40 minutes' exposure a t 30.5 cm. (1 foot) distance from this arc is equivalent, in the change in color produced, to a 2-hour exposure to brightest summer sun, the samples in both cases being in glass bottles. This is a sufficiently long exposure to differentiate between stable and unstable gasolines, and after this period further drop in color is usually slow. For routine work, further acceleration and nearly as accurate results may be obtained by using an exposure of 15 minutes in a quartz container or of 25 minutes in one of Corex D. This test has been used in these laboratories for 3 years and has proved rapid and reproducible.
143
Acknowledgment The authors wish to express their thanks to Wayne L. Benedict for assistance in the experimental work.
Literature Cited (1) Beard, L. C., Jr., and Reiff, 0. M., IND.ENG.CHEM.,Anal. Ed. 3, 280 (1931). (2) Coblentz, W. W., Dorcas, M. J., and Hughes, C. W., Bur. Standards, Sci. Paper 539 (1926). (3) Dorcas, M. J., IND.ENG.CHEM.,22, 244 (1930). (4) Greider, C. E.,Zbid., 23, 508 (1931). (5) Hoeman, E. C., Oil Gas J.,32, No.17,18 (1933). (6) Mathews, J. H., and Dewey, L. H.,J . Phys. Chern., 17, 216 (1913). RECEIVED January 28, 1936.
Determination of AlphamAmy lase S. JOZSA and W. R. JOHNSTON, The Fleischmann Laboratories, New York,
D
N. Y.
In revising the method of Jozsa and Gore for the determination of liquefying power, the authors have improved the experimentechnic and standardized the liquefying curve. By introducing a new type of enzyme unit termed the liquefon, they have been to calculate accurate values for
minute a t zero time under the given experimental conditions. Since the rate at zero time is directly proportional to enzyme concentration, the number of liquefons per gram of preparaJozsa and Gore (6) have pubtion is an exact measure of the lished a method for the determination of the liquefying power of a l p h a - a m y l a s e content and malt diastase, based on viscosity also of the liquefying power of measurements of a remarkably alpha-amylaseactivities from the measured the preparation at zero time miformstarch paste, which enamounts of liquefied starch. The enzymeunder the specified conditions. ables rather accurate measurement of the amount of s t a r c h unit method of expression is of general T h e n a t u r e of t h e l i q u e f o n liquefied when abha-amylase acts u n i t a n d i t s applicability to applicability and has been successfully apon the starch paste. However, alpha-amylase have been preplied to the study of other enzymes. thev auulied their exDerimenta1 In addiviouslv discussed (4. resdts *rather inaccurLte1y when t i o n i o t h e new"method of they assumed that the liquefying power of a given malt was proportional to the amount of starch calculation the authors have made significant changes in exliquefied under the experimental conditions. This assumption perimental procedure which are outlined below. leads to erroneous activity values unless the experimental range is narrowly limited. Fletcher and Westwood (1)have suggested the Equipment use of a restricted range, but this is not all satisfactory. Other workers (7) have proposed a rather complete modification, but their method of calculating activity values is not sound. They The equipment is essentially that used in the original report values for the liquefying power of a pancreatin preparamethod ( B ) , but with the following changes: It is very imtion which deviate from one another as much as 10 per cent. portant to select precipitating jars of uniform dimensions, These deviations may be due to their method of calculation, since the stirring efficiency depends on the size of the jar. A which apparently involves an empirical determination of certain constants for a given preparation and then the extension of pipet should be chosen which has a time of drainage from these constants to any alpha-amylase preparation. This extenmark to mark of 165 to 190 seconds a t 21' C. for the specified sion is certainly not accurate because of wide deviations in the glycerol solution. A constant-temperature bath should be rate curves of various preparations, particularly when the subused and the water pumped through the jacket of the pipet. strate is largely converted. Their expression of liquefying power is also inaccurate, since it represents only an average value for the The equipment could be easily modified, but it is always rate of conversion of substrate and is not proportional to the necessary to use reproducible technic. actual enzyme content-that is, a doubling of enzyme content would not double the liquefying power of the preparation as Preparation of Standard Starch Paste measured by them. U R I N G t h e last few years several investigators have studied the l i q u e f y i n g a c t i o n of alphaamylase on gelatinized starch.
In view of the above, a fundamental revision of the method is certainly necessary. The authors believe that the revision herein proposed enables one to measure accurately the activity of an alpha-amylase preparation. In order to calculate the concentration of alpha-amylase in a given enzyme preparation from the amount of starch liquefied under the specified conditions, the authors have introduced a new enzyme unit, the "liquefon," defined as that amount of starch-liquefying enzyme which will convert the standard starch paste a t the rate of 25 mg. of dry starch per
In making up the starch paste it is important to add the starch as soon as the water boils, in order to avoid any considerable loss of water. The first stirring should be for about 1.5 minutes and the second for 2 minutes. In cooling the paste one should use a water bath a t about 20" C. Too rapid cooling produces an unstable paste. In the modified method sodium chloride is used to stabilize the enzyme infusions. Accordingly, a satisfactory paste, when mixed with 10 per cent of its weight of water containing 250 mg. of sodium chloride per 100 ml. and stirred for l minute, should
144
INDUSTRIAL AND ENGINEERING CHEMISTRY
give the same outflow time a t 21 O C., within 10 to 15 seconds, as the glycerol solution specified (6). The time of stirring will vary according to the efficiency of the stirrer, and each operator must adjust his conditions to obtain a standard paste. After stirring with salt solution, a satisfactory paste has an outflow time within 10 to 15
-I
q- \
throughout. Viscometers of the Stormer type were tried, but are unsuitable for this work because of the pronounced disintegrating action of the rotating cup. Figure 1 shows the relationship between the percentage of starch liquefied and the percentage decline in outflow time for a pipet conforming to the specifications and for the modified Ostwald viscometer. Each point on the standard curve r e p r e sents the average of sixteen measurements with various pipets and each point on the curve for the modified Ostwald viscometer the average of three measurements. This curve was plotted on large-size cross-section paper and the values of the variables were read from the graph. The data obtained are given in Table I. TABLEI. PERCENTAGE DECLINEIN VISCOSITY AS A FUNCTION OF TEE AMOUNT OF STARCH LIQUEIFIED
J
Viscosity Decline % 1 2 3 4 5 6 7 8 9
10
FIGURE1. STANDARD LIQUEFYING CURVE
seconds of that of the glycerol solution, but for accurate measurements, the initial viscosity of the p a s t e t h a t is, the viscosity of the paste after stirring with water or enzyme solution-should check the outflow time of the glycerol solution as closely as possible. Accordingly, a paste after stirring with 10 per cent of its weight of water containing 250 mg. of sodium chloride per 100 ml. has a correct initial viscosity if its outflow time checks that of the glycerol solution within 2 seconds. The starch paste before stirring with water is rarely constant enough in viscosity to enable the operator to get the correct initial viscosity at the first stirring, but the correct viscosity is easily attained. If the paste is too viscous after stirring for 1 minute with 10 per cent of dilute sodium chloride solution, it is only necessary to extend the stirring of a second blank a few seconds until a correct initial viscosity is obtained. Similar considerations apply in the case of a paste of low viscosity. The enzyme tests should be stirred for the same length of time that is determined for the blank.
Liquefying Curve The “liquefying curve” as described in the original publication (6) represents the fundamental relation between viscosity and degree of liquefaction of the starch paste. Strictly, each pipet has its characteristic curve, but the authors have found that pipets of widely different outflow times gave essentially the same liquefying curve within 4 to 5 per cent deviation from an average curve. All the liquefying curves determined were smooth from beginning to end, which is not in agreement with the results of Fletcher and Westwood (I), who found the liquefying curve to be characterized by a rapid change in slope a t about 5 per cent liquefaction. Their results are possibly due to insufficient mixing. The liquefied starch must be gently but thoroughly mixed into the starch paste, particularly when the percentage of liquefied starch is small-below 10 per cent liquefaction. In addition, the pipet must be carefully cleaned and dried before each measurement if the greatest accuracy is to be realized. The authors also checked the liquefying curve by using a modified Ostwald viscometer and found that it was perfectly smooth
VOL. 7 , NO. 3
11 12 18 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
Starch Liquefied
%
Mg.
0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 3.0
13 25 38 51 63 76 89 101 114 127
Viscosity
Decline %
Starch Liquefied Mg.
%
51 52 53 64 65 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
so 9.7 10.1 10.5 10.9 11.3 11.7
494 ~. -
12.5 13.0 13.5
511 528 549 570
i5.i
409 426 443 460 477
81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
The values of Table I should be applicable to any pipet satisfying the prescribed conditions, with an accuracy of approximately 5 per cent. If greater accuracy is desired it is necessary to determine the characteristic curve of the given pipet. In using the standardized data it is only necessary to determine the final outflow time of the pipet. A fully liquefied starch is prepared according to the directions of the original method and its outflow time is recorded as the final outflow time of the particular pipet used. From the final and the initial outflow times the total range is obtained for a given pipet and the per cent decline in outflow time is calculated for a given measurement. The milligrams of starch liquefied are then read from the table. In order to have available a more complete evaluation of the standard curve, the authors derived the following equation which fits the curve with an accuracy of 1 to 2 per cent over the range from 0.0 to 90.0 per cent decline in outflow time:
MAY 15, 1935
ANALYTICAL EDITION
145
TABLTJ11. STARCH LIQUEFIED Decline
Liquefona per 10 00.
% 50.0 50.2 50.4 50.8 50.8 51.0 51.2 51.4 51.8 51.8 52.0 52.2 52.4 52.8 52.8 53.0 53.2 53.4 53.8 53.8 54.0 54.2 54.4 54.8 54.8 55.0 55.2 55.4 55.8 55.8 56.0 58.2 58.4 58.8 58.8 57.0 57.2 57.4 57.8 57.8 58.0 58.2 58.4 58.8 58.8. 59.0 59.2 59.4 59.6 59.8
0.891 0.898 0.700 0.705 0.709 0.714 0.719 0.725 0.730 0.735 0.742 0.748 0.751 0.758 0.761 0.788 0.771 0.777 0.782 0.787 0.792 0.799 0.805 0.811 0.817 0.824 0.830 0.838 0.843 0.849 0.856 0.862 0.889 0.878 0.883 0.889 0.898 0.903 0.910 0.917 0.924 0.931 0.938 0.948 0.953 0.981 0.988 0.975 0.983 0.990
Starch Mg. 794 799 804 809 814 819 826 831 838 842 848 853 858 883 888 873 878 883 889 894 899 905 911 917 923 929 936 941 948 952 958 984 970 978 982 988 994 1000 1005 1011 1017 1023 1029 1035 1041 1047 1053 1059 1064 1070
Decline
Liquefons per 10 cc.
% 80.0 60.2 80.4 80.8 60.8 61.0 81.2 61.4 81.8 61.8 62.0 82.2 62.4 62.6 62.8 63.0 83.2 83.4 83.6 83.8 64.0 64.2 84.4 84.8 84.8 65.0 85.2 85.4 65.6 85.8 66.0 68.2 86.4 68.6 68.8 87.0 87.2 67.4 87.8 87.8 88.0 88.2 88.4 88.8 88.8 89.0 89.2 89.4 69.6 69.8
0.997 1.006 1.015 1.024 1.033 1.043 1.052 1.061 1.071 1.080 1.090 1.100 1.109 1.119 1.129 1.139 1.149 1.159 1.169 1.179 1.189 1.201 1.213 1.225 1.237 1.249 1.261 1.274 1.287 1.300 1.313 1.327 1.339 1.352 1.360 1.379 1.393 1.407 1.420 1.434 1.446 1.463 1.478 1.493 1.507 1.522 1.538 1.553 1.568 1,584
Starch Mg. 1076 1083 1090 1098 1103 1110 1117 1124 1130 1137 1144 1151 1158 1164 1171 1178 1185 1191 1198 1204 1211 1219 1226 1234 1241 1249 1257 1284 1272 1279 1287 1295 1302 1310 1317 1325 1333 1340 1348 1355 1363 1371 1378 1388 1393 1401 1409 1418 1424 1431
+
S = 12.9P - 0.065P2 0.0025P3 where S = milligrams of starch liquefied P = percentage decline in outflow time
The standardization of the liquefying curve simplifies the method considerably and makes it more available to technological workers. The variable products of reaction produced by different amylase systems have no effect on the general a p plicability of the liquefying curve. This is shown by the fact that, for a given pipet, solutions having equivalent concentrations of liquefied starch, dextrins, and maltose have outflow times which deviate from one another by not more than 0.5 second. Evidence of this nature has also been submitted by Fletcher and Westwood (1).
Use of Sodium Chloride Jozsa and Gore (2, 8, 6)have shown that sodium chloride is effective as a desorption or activating agent and that it tends to stabilize enzyme infusions. Accordingly, the BUthors have used sufficient sodium chloride in preparing their infusions to make a concentration of 25 mg. per 10 ml. of infusion. In addition to its effect on the enzyme, sodium chloride decreases the viscosity of the starch paste. This effect makes necessary the use of sodium chloride in the blank determination.
Procedure The actual determination should be carried out as follows: The 150-gramsample of starch paste is cooled to about 19.5" C., so that after stirring in the enzyme infusion or the sodi2m chloride solution, the temperature of the stirred paste is 21 * 0.2' C.
Decline
Liquefona per 10 cc.
% 70.0 70.2 70.4 70.8 70.8 71.0 71.2 71.4 71.8 71.8 72.0 72.2 72.4 72.6 72.8 73.0 73.2 73.4 73.8 73.8 74.0 74.2 74.4 74.8 74.8 75.0 75.2 75.4 75.8 75.8 76.0 78.2 76.4 76.8 78.8 77.0 77.2 77.4 77.8 77.8 78.0 78.2 78.4 78.8 78.8 79.0 79.2 79.4 79.8 79.8
1.601 1.818 1.837 1.858 1.874 1.890 1.709 1.728 1.747 1.788 1,785 1.808 1.828 1.846 1.868 1.887 1.906 1.933 1.958 1.979 2.003 2.028 2.049 2.072 2.095 2.117 2,143 2.189 2.195 2.222 2.248 2.279 2.310 2.341 2.372 2.404 2.435 2.487 2.499 2.532 2.565 2.597 2.830 2.662 2.694 2.727 2.768 2.808 2.847 2.888
Starch Mg. 1439 1448 1458 1465 1473 1482 1490 1499 1507 1516 1524 1532 1541 1549 1558 1566 1575 1584 1594 1803 1812 1821 1629 1838 1848 1855 1884 1873 1683 1892 1701 1711 1721 1732 1742 1752 1782 1772 1782 1792 1802 1811 1821 1830 1840 1849 1880 1871 1882 1893
Decline
Liquefona per 10 cc.
% 80.0 80.2 80.4 80.6 80.8 81.0 81.2 81.4 81.8 81.8 82.0 82.2 82.4 82.8 82.8 83.0 83.2 83.4 83.8 83.8 84.0 84.2 84.4 84.8 84.8 85.0 85.2 86.4 85.8 85.8 86.0 88.2 88.4 88.8 86.8 87.0 87.2 87.4 87.8 87.8 88.0 88.2 88.4 88.8 88.8 89.0 89.2 89.4 89.8 89.8 90.0
2.929 2.971 3.014 3.057 3.102 3.148 3.191 3.238 3.282 3.329 3.375 3.429 3.482 3.537 3.593 3.850 3.705 3.783 3.822 3.880 3.940 4.005 4.071 4.139 4.208 4.277 4.347 4.419 4.493 4 567 4.842 4.725 4.810 4.896 4.982 5.072 5.166 5.283 5.361 5.460 5.582 5.668 5.775 5.883 5.998 8.108 8.231 8.355 8.482 8.812 8.744 I
Starch Mg. 1904 1915 1928 1937 1948 1959 1970 1981 1991 2002 2013 2025 2037 2049 2081 2073 2085 2097 2108 2120 2132 2145 2157 2170 2182 2195 2208 2220 2233 2245 2258 2272 2285 2299 2312 2328 2340 2354 2389 2383 2397 2411 2428 2440 2455 2489 2484 2499 2515 2530 2545
The degree of precooling will wry with the stirrer and surroundings, but in any case the paste after stirring should be at 21' C. The correct time of stirring for the proper initial outflow time is determined by running one or more blanks. Using this correct time, 15 ml. of enzyme infusion are stirred into 150 grams of paste and the mixture is placed in the bath at 21' C. After 59 minutes the mixture is sucked into the pipet and its outflow time determined. The measurement of the outflow time of the mixture is begun just before the end of the hour reaction period in order to correct for the liquefaction occurring during the measurement. In order to check the stability of the aste, another blank should be run on 150 grams of paste which gas stood for 1 or 2 hours at 21' C. The outflow time of this check blank should not deviate more than 3 or 4 per cent from the first blank. In pipetting the 15-ml. portions it is necessary to avoid the introduction of traces of saliva. A small cotton plug effectively prevents contamination.
Calculations From the outflow time of B given mixture the percentage decline is calculated, and from this the amount of starch liquefied is obtained by reference to the table or equation. The enzyme content or activity of the infusion is derived from the amount of liquefied starch by means of the empirical equation: (S - 1078) (0.000565) Loglo L where L = liquefons per 10 ml. of infusion and S = milligrams of starch liquefied in 1 hour
From the concentration of the infusion the number of liquefons per gram of preparation is calculated. The number of liquefons per gram of preparation is an exact measure of the alpha-amylase content and also of the liquefying power a t zero time.
INDUSTRIAL AND ENGINEERING CHEMISTRY
I46
An example of the calculation is given below:
Paste stirred with water for 60 seconds Additional stirring 10 seconds (trial) New sample stirred with water for 65 seconds (Outflow time with glycerol solution for this pipet is 170 seconds) Initial outflow time Final outflow time
Outflow Time Sec. 181.5 164.5 171.2
171.2
VOL. 7, NO. 3
Various amounts of different enzyme preparations must be used in order to obtain a liquefaction within 50 to 90 per cent decline in viscosity after 1 hour a t 21’ C. This range has been chosen as most suitable for accurate measurements. Table I11 gives the concentration of enzymic materials of various strengths in milligrams per 10 ml. necessary for proper application of the method.
56.5 ___
Ranen -..~ ~ _ . Outflow time with sample (10 mg. per 10 ml.) after 1 hour at 21’ C. Decline in outflow time: 171.2 - 91.8 = 79.4 seconds 7940 Per cent decline: 114.7 6g’2
114.7
TABLE111. ENZYMIC MATERIAL REQUIRED FOR ANALYSIS OF DIFFERENT PREPARATIONS
91.5
Liquefons per Gram
Enzymic Material per 10 MI. Infusion
1-3 2-6 5-15 10-30 20-60 50-150 100-300 200-600 500-1500 1000-3000
1000 600 200 100 60 20 10 5 2 1
-
Mg.
Mg. of starch liquefied (Table 11): 1409 Loglo L = (S - 1078) X (0.000565) where S = mg. of starch liquefied L = liquefons per 10 ml. 331 X 0.000565 = 0.1570 LogioL L = 1.538
Since 10 mg. of the sample were used and this figure must be related to 1 gram of the sample, the result should be multiplied by 100. The alpha-amylase concentration of the sample is measured by the liquefon content, which is 154 liquefons per gram. Its liquefying power a t zero time is 3850 mg. of dry starch per minute. Table I1 gives a convenient tabulation of data obtained from the liquefon-starch equation for the experimental range of 50 to 90 per cent decline in viscosity.
Preparation of Enzyme Solution Using barley malt (diastatic or pale), 5 grams of malt are weighed into a 1000-ml. flask, 26 grams of sodium chloride are added, and the flask is filled up to the mark, After standing for 1 hour at room temperature with occasional shaking, it is filtered, rejecting the fist 100 ml. of filtrate, 100 ml. of the filtrate are transferred into a 1000-ml. flask (or 60 ml. into a 500-ml. flask), and the flask is filled up to the mark. In case of diastatic malt sirup 2, 6, 10, or 15 grams should be weighed (accordingto strength) and 25 grams of sodium chloride added, in the same manner as in the above sample.
Acknowledgment The authors wish to express their thanks to C. N. Frey, H. C. Gore, and Q. Landis for their valuable suggestions and keen interest in this work.
Literature Cited (1) (2) (3) (4) (5)
Fletcher, L., a n d Westwood, J. B., J. Inst. Brewing, 36,550 (1930) Gore, H. C., and Jozsa, S., IND.ENG.CHEM., 24,102 (1932).
Ibid., 24, 99 (1932).
Johnston, W . R., and Jozsa, S., J. A m . Chem. SOC., 57, 701 (1935). Joasa, S., a n d Gore, H. C., IND.ENQ.CHEM.,24,95 (1932). (8) Ibid., Anal. Ed., 2,28 (1930). (7) Willaman, J. J., Clark, E. W., a n d Hager, 0. B., Biochem. Z . ,
258,94 (1933). RncmIvmD May 28, 1934. Presented before the Division of Biologics Chemistry a t the 86th Meeting of the American Chemical Society, Chicago, Ill., September 10 to 15,1933.
A Simple and Sensitive Test for p-Phenylenediamine OSCAR HEIM, 112-35 Dillon St., Jamaica, L. I., N. Y.
A
CCORDING to Erdmann (S), themost sensitive test for p-phenylenediamine is the white precipitate of quinone dichlorodiimide produced with sodium hypochlorite, and he recommends diazotation and coupling with &naphthol disulfonic acid (2) for its detection in fur. There are a number of other tests (1, C), but all have been found unsuitable for detecting the small residual quantities of p-phenylenediamine occasionally present in black dyed fur. The basis of the present test, originally intended to discover small amounts of p-phenylenediamine in fur which could not be found by tests commonly found and recommended for this purpose, is the formation of indamine when a mixture of pphenylenediamine and aniline is oxidized.
* \
0 0 +
N&
NHz
0 N/I
oxidation
’
I&
1
I
tion (1 drop of aniline in 50 cc. of water), mix, and add a few cry% tals of potassium persulfate. The appearance of a blue-green color within about 5 seconds indicates the presence of p henylenediamine or its derivatives in which the (toxic) p-$amine structure has been preserved-for example, p-toluenediamine sulfate, which can be considered the salt of a methyl-substituted pphenylenediamine, or dimethyl-pphenylenediamine. The test is a sensitive one: 0.05 cc. of a 0.00001 per cent aqueous solution (equivalent to 0.0008 mg. or 0.5 gamma of pphenylenediamine) will give a distinct coloration. Interfering substances do not present difficulties. Other amido compounds-so-called oxidation dyes which are employed in pelt dyeing, such as amido phenols, m-amido anisol, 0- and mphenylenediamine-do not respond to the test, but if a trace of p-phenylenediamine is present with these substances it will be detected.
Literature Cited NHz
Treat the test specimen with a small amount of a 3 per cent solution of acetic acid-for example,6.452 sq. om. (1 square inch) of fur with 1 to 2 cc. of acid-warm to about 45’ C., squeeze the liquid out into a porcelain crucible, add 1 drop of an aniline solu-
(1) Allen, A. H., “Commercial Organic Analysis,” 5th ed., Philadelphia, P. Blakiston’s Son & Co., 1925. (2) E r d m a n n , 2.angew. Chem., 18, 1377 (1905). ( 3 ) Ibid., 19, 1053 (1908). (4) Rosenthaler, L., “Der Nachweis organischer Verbindungen,” 2nd ed., Stuttgart, Verlag von Ferdinand Enke, 1923. RECEIYED March 26, 1935.
